In step one, we fit a suite of candidate models with different λ values on the training data, as described in Section 2.3.1.
Given our hypothesis that two saturable pathways governed benzene metabolism rather than one, we fit two Michaelis-Menten-like models to the data, one having a single metabolic pathway and the other having two pathways that competed for access to benzene (X).
In step one, we fit CAD status (with four levels: NO CAD, CAD, AMI and OLD MI) as the biological variable of interest, and adjusted for BMI, gender, ethnicity (in the discovery phase only since all individuals in the replication phase were Caucasian) and age (in the replication phase only, since it was not correlated with the major components of variance in the discovery phase).
As the accuracy of the estimate of the EC50 is greater when replicate observations are combined in one analysis (OECD 2004), we fitted one concentration response curve for each population × treatment combination (original, control and carbaryl-selected for each of the seven populations) using the results of all replicate trials with the different clones of this population.
To the scores of the twin 1 and 2 samples on these subtests, we fitted a one-factor model representing the general intelligence factor.
To explore the interaction further, for each bird for whom we had more than one escape flight, we fitted a linear regression model of take-off angle on take-off speed (see Fig. 3).
We need to get the ones we have fit.
Hence, beforehand, we decided to distinguish a class with change and one without, and we fitted the trend in the stable class with an intercept-only predictor.
Second, we fitted one-pollutant models, and then considered two-pollutant models by fitting one traffic-related and one stationary fossil fuel combustion-related pollutant.
